Purpose:We examined the validity and reproducibility of a squash-specifc test designed to assess change-of-direction speed.Methods:10 male squash and 10 male association-football and rugby-union players completed the Illinois agility run (IAR) and a squash change-of-direction-speed test (SCODS) on separate days. Tests were repeated after 24 h to assess reproducibility. The best time from three attempts was recorded in each trial.Results:Performance times on the IAR (TE 0.27 s, 1.8%, 90% CI 0.21 to 0.37 s; LOA -0.12 s ± 0.74; LPR slope 1, intercept -2.8) and SCODS (TE 0.18 s, 1.5%, 90% CI 0.14 to 0.24 s; LOA 0.05 s ± 0.49; LPR slope 0.95, intercept 0.5) were reproducible. There were no statistically significant differences in performance time between squash (14.75 ± 0.66 s) and nonsquash players (14.79 ± 0.41 s) on the IAR. Squash players (10.90 ± 0.44 s) outperformed nonsquash players (12.20 ± 0.34 s) on the SCODS (P < .01). Squash player rank significantly correlated with SCODS performance time (Spearman’s ρ = 0.77, P < .01), but not IAR performance time (Spearman’s ρ = 0.43, P = .21).Conclusions:The results suggest that the SCODS test is a better measure of sport-specific capability than an equivalent nonspecific field test and that it is a valid and reliable tool for talent identification and athlete tracking.